UNASSIGNED:建立并验证了社区获得性压力伤害(CAPI)的预测模型,以允许早期识别家庭护理人员和社区工作者的压力伤害风险。
UNASSIGNED:参与者是来自中国三级医院两个分院的65岁及以上的住院患者,一个用于模型训练集,另一个用于验证集。本研究是基于医院电子病历的病例对照研究。根据入院时压力伤的存在,将患者分为病例组和对照组。在模型训练集中,LASSO回归用于选择最佳预测因子,然后使用逻辑回归构建列线图。通过绘制受试者工作特性曲线(ROC)并计算曲线下面积(AUC)来评估模型的性能,校准分析,和决策曲线分析。该模型使用10倍交叉进行内部和外部验证。
未经评估:该研究共包括20,235名受试者,包括训练集中的11,567和验证集中的8668。CAPI在训练集和验证集中的患病率分别为2.5%和1.8%,分别。列线图包括八个变量:年龄≥80岁,营养不良状况,脑血管意外,低蛋白血症,呼吸衰竭,恶性肿瘤,截瘫/偏瘫,和痴呆症。原模型中预测模型的AUC,内部验证,外部验证为0.868(95%CI:0.847,0.890),平均0.867和0.840(95%CI:0.807,03.873),分别。列线图显示可接受的校准和临床益处。
UNASSIGNED:我们构建了一个列线图,以从合并症的角度预测CAPI,该合并症适合非专业人士使用。此列线图将帮助家庭护理人员和社区工作者早期识别PI风险。
UNASSIGNED: A predictive model of community-acquired pressure injury (CAPI) was established and validated to allow the early identification of the risk of pressure injuries by family caregivers and community workers.
UNASSIGNED: The participants were hospitalized patients 65 years and older from two branches of a tertiary hospital in
China, one for model training set and the other for validation set. This study was a case-control study based on hospital electronic medical records. According to the presence of pressure injury at admission, patients were divided into a case group and a control group. In the model training set, LASSO regression was used to select the best predictors, and then logistic regression was used to construct a nomogram. The performance of the model was evaluated by drawing the receiver operating characteristic curve (ROC) and calculating the area under the curve (AUC), calibration analysis, and decision curve analysis. The model used a 10-fold crossover for internal and external validation.
UNASSIGNED: The study included a total of 20,235 subjects, including 11,567 in the training set and 8668 in the validation set. The prevalence of CAPI in the training and validation sets was 2.5% and 1.8%, respectively. A nomogram was constructed including eight variables: age ≥ 80, malnutrition status, cerebrovascular accidents, hypoproteinemia, respiratory failure, malignant tumor, paraplegia/hemiplegia, and dementia. The AUC of the prediction model in the original model, internal validation, and external validation were 0.868 (95% CI: 0.847, 0.890), mean 0.867, and 0.840 (95% CI: 0.807,03.873), respectively. The nomogram showed acceptable calibration and clinical benefit.
UNASSIGNED: We constructed a nomogram to predict CAPI from the perspective of comorbidity that is suitable for use by non-specialists. This nomogram will help family caregivers and community workers with the early identification of PI risks.